VOGONS


First post, by Almoststew1990

User metadata
Rank Oldbie
Rank
Oldbie

Hello everyone,

I have just watched Phil's video on using an SD Card with Windows 98. I am tempted to try it out but I have a couple of questions!

If my IDE controller can do speeds up to 33MB/s, the SD-IDE adapter can do 24MB/s, would a Class 4 30MB/s SD Card be fast enough to NOT be the bottleneck? Is 30MB/s the 'theoretical' max speed, and in most usage it will only manage 10MB/s?

I have an XP PC and I am tempted to use this as it would make the system neat and tidy! My XP PC uses a cheap HDD over SATAII and gives performance like this or worse (the benchmark is from when the HDD was in my SATAIII W10 PC):

9MTSyUv.png

Would a 24MB/s SD card give better real-world performance than a 80MB/s SATA HDD, given that access time is much better on the SD Card?

What sort of access time does an SD card get in comparison to the 17ms of my HDD?

Ryzen 3700X | 16GB 3600MHz RAM | AMD 6800XT | 2Tb NVME SSD | Windows 10
AMD DX2-80 | 16MB RAM | STB LIghtspeed 128 | AWE32 CT3910
I have a vacancy for a main Windows 98 PC

Reply 1 of 14, by PhilsComputerLab

User metadata
Rank Hardware Mod
Rank
Hardware Mod

Use something like ATTO benchmark to get a better picture.

You want to use the best SD card you can, regardless of transfer rate.

I really should do a video so that people can stop obsess about transfer rate. Access time and especially performance with tiny files is much more important. And CPU performance is the other factor.

Good SD cards don't cost a fortune, something like a Samsung EVO for example. There are tons of benchmarks on camera, and other sites, but if you buy a generic class 10 card even, they can be noticeably slower than a decent branded one. They all satisfy Class 10, but performance with small files can vary massively.

To answer your question, yes, a half decent SD card wipes the floor with an old 80 GB hard drive, in a machine that can take advantage of it. E.g. don't expect your Pentium 133 to be much faster...

YouTube, Facebook, Website

Reply 2 of 14, by zerker

User metadata
Rank Member
Rank
Member

I'm not sure where you're getting your numbers for the SD cards, but the class number is only the guaranteed sustained WRITE speed, in megabytes per second. E.g. a class 4 is guaranteed to give you 4 MB/s. However, like all flash memory, that will degrade with small files writes as Phil said.

I don't believe the speed classes make ANY guarantee on read speeds at all. You'd have to benchmark your individual card in a suitably fast (e.g. USB3) card reader to get a good estimate of read speeds. I second Phil's suggestion of using ATTO Disk Benchmark, which checks both read and write performance; your screenshot doesn't seem to differentiate.

... though I'd like to find something that runs directly on Win 98 for comparisons 😀.

The benchmarks I got from my two Kingston Class 4 cards sitting around were between 16 and 24 MB/s read speeds (depending on capacity) and write speeds actually exceed specs and were about 12 MB/s. In both cases, performance dropped off substantially at less than 64k write/read sizes. The mechcanical drives I benchmarked, for the record, only dropped off below the 8 Kb mark.

I did a benchmark of my 1 GB CF card on my retro PC a couple days ago using speedsys, in DOS. It came up with:
Random access time: 0.69 ms
Buffered read speed: 6648 KB/s
Linear verify speed: 36968 KB/s
Linear read speed: 6648 KB/s

But I'm not entirely sure what the verify speed is, and it doesn't seem to have write tests. However, that should show you the access times on flash media are really, really low.

So yes, there are a lot of variables and real world performance will vary considerably. I would personally look for something with a bit better write performance, but honestly, you can always try it, then image your install onto a faster card if you aren't satisfied.

I've uploaded a pile of benchmarks on all my disks on modern hardware if it's useful to you:
http://www.zerker.ca/misc/Drive%20Benchmarks.ods

You can see if those numbers are any use to you. It's a jumble of USB, CF, SD, SSD, Mechanical and even Memory Sticks, but I've tried to make all the labels clear as to what card and what class each one is. They were all benchmarked using ATTO and a quick Python script I cooked up to grab the values out of the proprietary result file format.

Last edited by zerker on 2017-05-07, 12:41. Edited 1 time in total.

Reply 3 of 14, by PhilsComputerLab

User metadata
Rank Hardware Mod
Rank
Hardware Mod

This one runs under Windows 98. Been using it for a long time 😀

Attachments

  • Filename
    atto_disk_benchmark_241.zip
    File size
    232.97 KiB
    Downloads
    23 downloads
    File comment
    Runs under Windows 98 :D
    File license
    Fair use/fair dealing exception

YouTube, Facebook, Website

Reply 4 of 14, by zerker

User metadata
Rank Member
Rank
Member

Thanks Phil. I went ahead and benchmarked my CF card running directly on Windows 98 for additional comparison. Here's a rough idea of how the raw original performance compares when running over the IDE adaptor:

cfcompare.png

Spreadsheet with original numbers can be found at:
http://www.zerker.ca/misc/CF%20Adaptor%20Compare.ods

NOTE that this is a pretty good Transcend Industrial 16 GB CF card. I.g. exactly the sort of card designed for use as a 'hard drive'. Lowest small-file performance is still 2-3 MB/s for both reads and writes, which is pretty damn good. I don't have a single SD card (even my speedy U3 rated cards) that comes close to that performance at the small end. To be fair, my Sandisk Extreme 8 GB CF card is similarly hobbled. I guess that explains why it took FOREVER to put Linux on that thing.

I also edited and re-uploaded my script to support the old ATTO benchmark format too 😀

.. I don't suppose you also have good suggestions for Win 3.1/DOS DISK benchmarking too?

Last edited by zerker on 2017-05-07, 13:27. Edited 1 time in total.

Reply 5 of 14, by PhilsComputerLab

User metadata
Rank Hardware Mod
Rank
Hardware Mod

Nah got nothing for DOS I'm afraid. IMO DOS doesn't need very fast storage to be responsive 😀

I guess for Windows 3 you do though.

YouTube, Facebook, Website

Reply 6 of 14, by clueless1

User metadata
Rank l33t
Rank
l33t

speedsys does have some storage benchmarks built in. (for DOS)

The more I learn, the more I realize how much I don't know.
OPL3 FM vs. Roland MT-32 vs. General MIDI DOS Game Comparison
Let's benchmark our systems with cache disabled
DOS PCI Graphics Card Benchmarks

Reply 8 of 14, by Almoststew1990

User metadata
Rank Oldbie
Rank
Oldbie

I thought i'd bump an old thread that came up in a Google search instead of starting a new one (which turned out to be my own bloody old thread!) about what SD card to get. After this post in 2017 I got CF cards and they've been great but they're pretty expensive for anything bigger than 8GB.

For SD cards, can I get a 64GB SD Card for use as a C: drive? Does it need to be a certain type to work? This will be for Windows 98 and XP.

Ryzen 3700X | 16GB 3600MHz RAM | AMD 6800XT | 2Tb NVME SSD | Windows 10
AMD DX2-80 | 16MB RAM | STB LIghtspeed 128 | AWE32 CT3910
I have a vacancy for a main Windows 98 PC

Reply 9 of 14, by debs3759

User metadata
Rank Oldbie
Rank
Oldbie

I don't see why not, as long as you have the right adapter. They're not expensive, lots on ebay for under £16, so probably even cheaper if you look around.

See my graphics card database at www.gpuzoo.com
Constantly being worked on. Feel free to message me with any corrections or details of cards you would like me to research and add.

Reply 10 of 14, by douglar

User metadata
Rank Member
Rank
Member

SD cards with an “Application Performance Class” speed rating should have the best performance profile for retro computing. https://www.sdcard.org/developers/overview/ap … tion/index.html

But some people have not seen a big advantsge with the A rated cards. https://www.jeffgeerling.com/blog/2019/raspbe … -fools-me-twice This is especially true if you can’t get the cache and command queue features to work with your adapter.

Grey market devices on ebay or amazon can be a waste of money. Those $10 wonders have left me wondering things like “how can this SD be so slow?” Check to make sure you are getting a real product, and not a cheap knockoff with a bad paint job. https://www.diyphotography.net/psa-fake-sandi … cluding-amazon/

For win98, set ConservativeSwapfileUsage=1 in your system ini to prevent aggressive memory paging from aging your SD prematurely.

Reply 11 of 14, by jmarsh

User metadata
Rank Oldbie
Rank
Oldbie

For SD (instead of CF) the adapter you get is more important than the card. Cheap adapters will be terribly slow and in worst cases may not present the card as a consistent device i.e. the number of cyls/heads/sector values change between power cycles, or some other part of the ATA IDENTIFY response changes to make windows think it's a different drive.

Reply 12 of 14, by kolderman

User metadata
Rank Oldbie
Rank
Oldbie
jmarsh wrote on 2020-09-24, 04:03:

For SD (instead of CF) the adapter you get is more important than the card. Cheap adapters will be terribly slow and in worst cases may not present the card as a consistent device i.e. the number of cyls/heads/sector values change between power cycles, or some other part of the ATA IDENTIFY response changes to make windows think it's a different drive.

Cheap ones may not support DMA either. CF is best for DOS by far.

Reply 13 of 14, by Almoststew1990

User metadata
Rank Oldbie
Rank
Oldbie

On UK eBay, once I filter out the thousands of results from China, there are 7 SD to IDE adapters available. Only 2 use a 40 pin rather then 44 pin connector.

https://www.ebay.co.uk/itm/123368184975

https://www.ebay.co.uk/itm/333271363033

Would either of those be OK?

The unbranded ones seem to be a few pounds less or about 8 if I order direct from China, but otherwise look identical.

Ryzen 3700X | 16GB 3600MHz RAM | AMD 6800XT | 2Tb NVME SSD | Windows 10
AMD DX2-80 | 16MB RAM | STB LIghtspeed 128 | AWE32 CT3910
I have a vacancy for a main Windows 98 PC

Reply 14 of 14, by douglar

User metadata
Rank Member
Rank
Member

The links you posted look like Sintech clones. I've used a couple the Sintech clones that usually have a chip labeled FC1307A chip. They are most common. My experience with them has been mixed.

Pros:

  • Cost effective for > 8GB volumes
  • Most techies have a collection of old small SD cards in a drawer, ready for re-use, making them cost effective for small volumes too
  • Easier to swap between computers than a 40 pin DOM (designed to fit tightly) or CF devices (I managed to bend pins in the adapter)
  • Good performance with read heavy, random access work loads, making it a great performer in DOS
  • Max speed is more than fast enough for anything before a 486/66
  • SD's should have more than enough endurance for any DOS / Win3.1 work load

Cons:

  • Compatibility with mid/late 90's E-IDE controllers was less than desired. Not uncommon for ATA-3 through UDMA-4 controllers to fail to identify devices or negotiate an ATA protocol slower than desired.
  • Can be slow if the device falls back to ATA-0 for compatibility, maxing out at 2MB/s effective max throughput, which can be noticeable on a pentium with an Eide adapter.
  • No slave/master jumper.
  • Sinitech devices sometimes appear to have an internal max transfer rate of 25MB/s which will limit a fast SD on a UDMA-6 controller
  • Seems like the lack of wear leveling & trim could shorten the SD life on a frequently used WinXP install.
  • Sintech is tough to mount and has exposed power pins on rear, which scares a clumsy person like myself

My experience with IDE "DOM" devices and CF devices was more pleasant. Less dangling cable anxiety. I didn't see any compatibility issues after playing around with a many different DOMS and CF cards on the same hardware that didn't like the Sinitech. and when all is said and done, the price for the DOM & CF devices was comparable to the Sinitech& SD when working with devices < 4GB .

There are other SD to IDE chips that are less common. Anyone have experience with those devices?

This guy did some tests: https://goughlui.com/2019/02/03/tested-generi … dapter-sd35vc0/